Procedure for the autonomous treatment of soil surfaces

Abstract

The invention relates to a method for the autonomous processing of floor surfaces by means of an autonomous mobile device, in particular a floor cleaning device such as a vacuum robot (4) and / or a sweeping and / or scrubbing robot, comprising the following steps: performing a scanning pass of the autonomous mobile device in a proposed floor processing area to establish a map of the environment (10), detecting obstacles (2, 3) by means of a detection device, classifying the obstacles (2, 3) as traversable or impassable, and moving over an obstacle (2) classified as traversable and / or around an obstacle (3) classified as impassable. The invention also relates to an autonomous mobile device suitable for carrying out said method.

Description

Procedure for the autonomous treatment of soil surfaces The invention relates to a method for the autonomous treatment of floor surfaces using a self-propelled mobile device, in particular a floor cleaning device such as a robotic vacuum cleaner, sweeper, and / or scrubber. The invention also relates to a self-propelled mobile device with which such treatment can be performed. Mobile, self-propelled devices, such as robotic vacuum cleaners, aim to autonomously clean an entire floor surface as thoroughly as possible. However, obstacles such as furniture, decorative items, small objects, or door thresholds prevent the robot vacuum from reaching all areas of the floor. Robot vacuum cleaners attempt to overcome thresholds and obstacles when they are in their path. Obstacles that are too large are usually detected by collision sensors mounted along the direction of movement. The robot vacuum cleaner can pass over small obstacles. However, beyond a certain height, the robot vacuum cleaner is unable to pass over them.After several attempts, the vacuum robot registers that it cannot overcome the obstacle and then tries to avoid that point, which may mean that the corresponding obstacle has already been damaged. Depending on the robot vacuum's capabilities, it can overcome small thresholds by simply stepping over them. For example, it can go over door thresholds to continue cleaning. The robot vacuum can overcome these with sufficient power. Other thresholds, such as floor-level crossbars on furniture like rocking chairs or clothesline legs, cannot be crossed by the robot vacuum. It must go around them. Low-profile thresholds pose a risk of the robot vacuum getting stuck after passing over them and being unable to free itself. Especially on furniture with valuable surfaces, such as chrome, varnish, or glass, the robot's attempts to free itself can unfortunately cause damage to the surface.Annoying and unpleasant noises may also occur during attempts to release the vacuum robot, which negatively impact the user's overall experience. Documents WO 2019 / 004742 A1, KR 20200094816 A and US 2020 / 125113 A1 describe, for example, vacuum robots that detect obstacles in their environment and, among other things, classify them. The objective of the invention is to provide a procedure for the autonomous treatment of floor surfaces that is effective, optimized and / or non-damaging, ensuring in particular the most complete cleaning possible of the floor surface, while preventing the self-propelled mobile apparatus from damaging obstacles. This objective is achieved by means of a procedure for treating soil surfaces with the characteristics of claim 1 and by means of a self-propelled mobile apparatus with the characteristics of claim 9. Advantageous improvements and refinements are the subject of the secondary claims. According to the invention, it includes a method for the autonomous treatment of floor surfaces with the aid of a self-propelled mobile device, in particular a floor cleaning device, such as a robotic vacuum cleaner and / or sweeper and / or scrubber, the following steps of the method: carrying out an exploratory drive of the motorized mobile device in a planned soil treatment area, in order to create a map of the surroundings, obstacle detection by means of a detector device, determining the position of the detected obstacles on the map of the environment by the mobile motor vehicle, classification of obstacles as surmountable or insurmountable and to overcome an obstacle classified as surmountable and / or detour around an obstacle classified as insurmountable, with obstacles close to the door being classified as the door threshold and obstacles far from the door as furniture. The solution corresponding to the invention is characterized by the classification of detected obstacles based on their position on the environmental map. Obstacles located in a passageway are classified as surmountable. Other obstacles are classified as insurmountable, and the self-propelled vehicle (SPV) circumvents them before attempting to cross them. This advantageously prevents damage to these obstacles. The SPV, therefore, assesses whether a detected obstacle is surmountable before attempting to cross it. The SPV then classifies the obstacles based on data contained in the environmental map.In particular, the mobile motor vehicle identifies an obstacle before traversing it, compares it with information from known map data, classifies it as surmountable or unsurmountable, and reacts accordingly, passing over obstacles classified as surmountable and / or going around obstacles classified as unsurmountable. Based on the solution provided by the invention, the risk of the self-propelled mobile unit becoming stuck on flat obstacles, such as chair supports at floor level, is advantageously reduced. The risk of damaging valuable furniture is also reduced. Furthermore, the wear and tear on the mobile unit is minimized due to reduced contact with the furniture. Advantageously, there is less noise during the cleaning process, as the unit no longer collides with flat obstacles, for example, at maximum speed, or becomes stuck, with its wheels rattling over the obstacle. By reducing the risk of the unit becoming stuck, the cleaning task is often completed advantageously without user intervention. The risk of the cleaning being interrupted is also advantageously reduced. Cleaning time can also be reduced, as the unit does not waste time attempting to overcome obstacles. A self-propelled mobile device is understood to be, in particular, a device for cleaning floors, such as a floor cleaner or lawnmower, that autonomously treats floor or lawn surfaces in a domestic setting. This includes robotic vacuum cleaners, sweepers, and / or scrubbers, such as robotic dust collectors or robotic lawnmowers. These devices operate in service (cleaning or mowing) preferably without or with minimal user intervention. For example, the device moves autonomously within a predetermined room to clean the floor according to a predetermined and programmed process strategy. A survey move, in particular, refers to a movement undertaken to gather information about a floor area to be treated, such as obstacles, room configuration, and similar factors. The objective of a survey move is specifically to estimate and / or represent the characteristics of the floor area to be treated. The term "area of ​​the floor to be treated" refers to any area of ​​the room that is intended to be treated, particularly cleaned. This could refer, for example, to a single room (within a dwelling) or an entire dwelling. It can also simply refer to areas within a room (of a dwelling) or a dwelling that are intended to be cleaned. Obstacles shall be understood to mean any objects or things located in the area of ​​the floor to be treated, which, for example, are there and which influence the treatment by the mobile self-propelled apparatus, in particular obstruct and / or disturb, such as thresholds, door thresholds, furniture, walls, curtains, carpets and the like. Overwhelmable obstacles are understood to be those obstacles that a motor vehicle can pass over due to its low height, without becoming stuck or colliding with them. For example, door thresholds or carpets can be classified as overwhelmable obstacles. Insurmountable obstacles are understood to be obstacles that the motor vehicle cannot pass over due to their height; that is, it would either run against them or get stuck there, thus interrupting the cleaning process. After the exploration phase, the mobile vehicle learns its surroundings and can transmit this information to the user as an environmental map, for example, via a mobile app. Detected obstacles that can be overcome and those that cannot be overcome are displayed prominently on the environmental map. Obstacles are given special prominence based on their classification. For example, obstacles classified as overcomeable are represented in a different color, shape, or similar way than obstacles classified as insurmountable. An environmental map should be understood as any map that is suitable for representing the surroundings of the area of ​​land to be treated, including all its obstacles. For example, the environmental map shows the area of ​​land to be treated, including the obstacles contained therein and the walls, in a sketch-like manner. The map of the environment, including obstacles, is preferably displayed in the app on an additional portable device. This is particularly useful for the user to visualize potential interactions. In this particular case, an additional device is understood to mean any device that a user can carry, external to the motor vehicle, specifically differentiated from the motor vehicle and suitable for displaying, providing, transmitting and / or transferring data, such as a handheld device, a smartphone, a tablet and / or a computer or laptop. The app, specifically a cleaning app, is installed on the portable unit. This app connects the mobile unit to the portable unit and allows the user to visualize the area to be cleaned, i.e., the room or area of ​​the house to be cleaned. The app then displays the area to be cleaned as a map of the surroundings, highlighting any obstacles. A detection device is understood to be any device suitable for detecting both surmountable and insurmountable obstacles, preferably with reliability. It is preferably based on lasers, sensors, and / or cameras. Classification refers specifically to the division of obstacles and / or objects into surmountable and insurmountable. Additional classifications may also be made, such as flat and uneven obstacles, or similar categories. In an advantageous embodiment, obstacle classification is performed based on existing map data obtained during exploration. Specifically, recognized obstacles are compared with information from the surrounding map to estimate and / or classify them accordingly. Preferably, classification is performed by comparing information from exploration with information obtained during obstacle detection. Ideally, the self-propelled mobile device autonomously identifies spaces for classification based on information from its exploration. For example, the mobile self-propelled device, having traveled during its exploration phase, uses a map of its surroundings. Based on the geometry, the device can estimate which areas actually correspond to which spaces. Narrow areas on the surrounding map, located between adjacent rooms, are identified as doorways or door thresholds and are therefore classified as passable. Alternatively, the user can use the app on the mobile device itself to specify which surfaces or areas on the surrounding map correspond to which rooms in their home. The device can also autonomously identify narrow spaces between adjacent rooms, such as doorways or thresholds, and classify them as passable. In an advantageous embodiment, obstacles are classified before an attempt is made to cross them. Therefore, in this case, no attempt is made to cross obstacles classified as insurmountable, to avoid potential damage to the obstacles or the vehicle becoming stuck. According to the invention, the self-propelled mobile device determines the position of detected obstacles on the environmental map. In particular, it determines, as accurately as possible, where the detected obstacle is located on the environmental map. If the obstacle is found in an area near a door or a transition between rooms, then the obstacle is classified as a threshold, specifically a door or room threshold, which can be crossed. If the detected obstacle is found inside a room, then it can be assumed to be a piece of furniture. This is surrounded by the self-propelled mobile device during cleaning. Therefore, in particular, according to the invention, obstacles close to the door and / or close to the wall are classified as door thresholds and obstacles far from the door and / or far from the wall as furniture, by going over the obstacles close to the door and going around obstacles far from the door, before making an attempt to go over the obstacles far from the door. According to the invention, it includes a self-propelled mobile device, in particular a floor cleaning device, for the autonomous treatment of floor surfaces such as a robotic vacuum cleaner and / or sweeper and / or scrubber, a detection unit for detecting obstacles, and an evaluation unit for classifying obstacles as surmountable or insurmountable. Specifically, the detection unit includes sensors that determine distance measurements and / or variations in sensor values ​​over time. Any feature, configuration variant, form of implementation and relative advantage of the procedure, also apply in relation to the corresponding self-propelled mobile apparatus of the invention and vice versa. An assessment unit is understood to mean any equipment suitable for classifying obstacles and / or objects as surmountable or insurmountable, particularly based on the obstacle's position within its environment. A more detailed classification of the obstacles is not strictly necessary, but may have been carried out. To detect an obstacle before the motor vehicle encounters it, a sensor system similar to the familiar cliff sensors can be used. When the distance measurement increases, it indicates that the motor vehicle is in a gap or has been lifted by the user. However, when the measurement falls below a certain threshold, it signifies that the obstacle has been reached. In addition to using the device-mounted cliff sensors, other mounting methods can be used to ensure reliable threshold detection. Elevated and / or oblique mounting positions of the sensors guarantee threshold detection before contact with the device. Forward-facing positions, or positions that can be elastically compressed, allow detection at short distances. Furthermore, the variation in sensor values ​​over time can be captured to differentiate between flat obstacles, such as rocking chair supports at floor level, flat table legs, or similar items, and carpets. When dealing with flat obstacles, a continuous profile can be detected in the evolution of sensor values ​​over time, while, conversely, a steep structure in carpets produces irregular sensor values. Alternatively, the obstacle can be detected by a second bumper positioned at the obstacle's height on the access ramp. The elastic force has then been chosen so that a soft obstacle, such as a carpet, is detected as such and can be distinguished from an obstacle like a chair crossbar. Using existing room map data corresponding to the environmental map, a threshold is reliably detected, allowing the device to be checked and classified as either a door threshold or an obstacle within the room. The device passes over door thresholds and around other obstacles. The invention will be described in more detail based on the following embodiments, which are merely examples. It is shown in: Figure 1: A schematic view of an example of an environmental map being created by performing the procedure corresponding to the invention for the automatic treatment of floor surfaces with the help of a self-propelled mobile device. Figure 2: A schematic detail view of the environment map of the embodiment of Figure 1, Figures 3A-3C: A schematic cross-section of an embodiment of a self-propelled mobile apparatus during the performance of the procedure corresponding to the invention for the automatic treatment of floor surfaces and Figure 4: A schematic flow diagram of the sequence of the procedure corresponding to the invention for the automatic treatment of floor surfaces. Figure 1 shows an environment map 10, generated by a self-propelled mobile device, specifically a vacuum robot, during a scanning operation. During this scanning operation, all obstacles, such as furniture, door thresholds, carpets, and curtains hanging within a predetermined floor treatment area, are detected by a sensor on the vacuum robot. Based on the geometry of the environment map, the vacuum robot can estimate which areas correspond to actual rooms. Specifically, individual boundaries on the environment map correspond to different sections of the room and / or to different rooms 1a-1i. Alternatively to the autonomous classification of rooms by the vacuuming robot, the user can prescribe, through an app such as on their mobile device, which boundaries of the environment map correspond to which rooms 1a-1i of their home. In addition to rooms, the vacuuming robot detects any obstacles within them. It then classifies narrower areas on the environment map between adjacent rooms as doorways, doorways, and / or door thresholds, and delimits these areas in relation to obstacles (3) located within a room. Therefore, when the vacuuming robot detects an obstacle (2, 3) ahead of it during its movement, it can adjust its course based on the obstacle's position on the environment map (10). If the obstacle is located near a door or between two adjacent rooms, it is classified as a door threshold, which it can overcome. If the obstacle is located within a room, it is classified as a piece of furniture. The vacuuming robot then moves around it during cleaning.An attempt to go over or overcome obstacle 3 is not made then, to avoid damage and / or the vacuum robot getting stuck. Specifically, the vacuuming robot classifies obstacles 2 and 3 as surmountable or insurmountable based on their position on the map of the environment. Obstacles near door 2 are classified as door thresholds, and obstacles far from door 3 are classified as furniture. As a result of this classification, the robot passes over obstacles near door 2 and goes around obstacles far from door 3 before attempting to pass over those far from door 3. Following the classification of obstacles 2 and 3, the robot passes over obstacles 2 classified as surmountable and goes around obstacles 3 classified as insurmountable. Figure 2 shows a detail of the environment map 10 from the implementation example in Figure 1. The vacuum robot has detected individual rooms 1c to 1g on the environment map and memorized them accordingly. Between individual rooms 1d and 1e, and 1c and 1d, the vacuum robot has detected obstacles 2 and classified them as door thresholds, based on their position in an intervening space. The vacuum robot can pass over these obstacles 2 near the door or near the wall during the cleaning program without damaging the obstacle or getting stuck. Obstacles 3 that, due to their position, are located within the room are classified as furniture.These obstacles, located far from the door or far from the wall, are circled by the vacuum robot during the cleaning program, without attempting to go over them, to avoid damage to obstacle 3 or the vacuum robot getting stuck. Figure 3 shows a side view of a vacuum robot 4, which detects an obstacle 2 as a door threshold and classifies it as such. To detect a door threshold before the vacuum robot 4 makes contact with it or passes over it, the robot includes at least one distance sensor 5, similar to well-known cliff sensors. When its distance measurement increases, the vacuum robot 4 is either in an opening or has been lifted by the user. Conversely, when the distance measurement decreases within a defined area by a certain amount, a door threshold is detected. This allows door thresholds to be detected before the vacuum robot makes physical contact with them or reaches them. Figure 3A shows the vacuum robot 4 moving in the direction of travel F and detecting obstacles along its laser beam 6 with its distance sensor 5. Figure 3B shows an alternative configuration of the distance sensor 5 on the vacuum robot 4. Figure 3C shows a retractable, front-mounted sensor 5 on the vacuum robot 4, enabling it to detect obstacles even at short distances. For this purpose, the sensor 5 is attached to a spring 7. The spring 7 can then retract in direction K, allowing the sensor 5 to be recessed into the vacuum robot, thus protecting it from impacts and damage when inserted. Additionally, the time-dependent measurement of distance values ​​from sensor 5 is preferentially captured to differentiate between smooth obstacles, such as rocking chair supports at ground level or flat table legs, and rough obstacles, such as carpets. With smooth obstacles, the time-dependent evolution of sensor values ​​results in a continuous profile, while with rough surfaces, irregular sensor values ​​are produced. Alternatively, the obstacle to be detected can be placed at the height of the obstacle on the access ramp (not shown). For this purpose, the bumper's elastic force has been chosen such that a soft object, such as a carpet, is detected as such and can be distinguished from an obstacle such as a chair crossbar. If a map of the environment with recognized rooms exists and the robot vacuum detects a doorway, it can then check and classify the robot vacuum based on whether it is a doorway or an obstacle within the room. Doorways are crossed by the robot vacuum, while other obstacles are bypassed by the robot vacuum. Figure 4 illustrates this flowchart. Initially, in step 11 of the procedure, the vacuum robot performs a scanning movement in a designated floor treatment area and creates a map of its surroundings. In step 12, the vacuum robot moves within its known environment, for example, to fulfill a cleaning assignment. The vacuum robot can access the environment map containing recognized or designated rooms. In step 13, the vacuum robot's sensor system detects an obstacle, such as a doorway, in front of it. The vacuum robot checks its environment map to see if the obstacle is located in a doorway zone (step 14). If the obstacle is located in an area between two rooms or in a user-cleared zone (path 15a), then the obstacle is classified as surmountable, for example, as a doorway.In procedure step 16a, the vacuum robot attempts to overcome the obstacle. If, on the other hand, the obstacle is located inside a room (path 15b), it is classified as insurmountable, for example, as a piece of furniture. The vacuum robot assesses the obstacle as such and looks for a way around it, without attempting to go over it (step 16b). After overcoming or going around the detected and classified obstacle, the vacuum robot continues its movement and cleaning task in procedure step 17. Based on the position or location of the detected obstacle on the environment map, the obstacle is classified, and the robot vacuum's next steps are determined accordingly. This significantly reduces the risk of the robot vacuum getting stuck on flat obstacles, such as chair legs at floor level. It also advantageously reduces the risk of damaging valuable furniture by running over it or attempting to do so. Furthermore, it minimizes wear and tear on the robot vacuum itself due to minimal direct contact with obstacles. This can result in lower noise levels and a reduced risk of getting stuck during cleaning. Overall, cleaning time can be significantly reduced, as the robot vacuum's attempts to overcome obstacles are prevented.

Claims

1. A method for the autonomous treatment of floor surfaces using a self-propelled mobile device, in particular a floor cleaning device such as a robotic vacuum cleaner (4) and / or sweeper and / or scrubber, comprising the following steps: performing a scanning movement of the self-propelled mobile device in a planned floor treatment area to create a map of the surroundings (10), detecting obstacles (2, 3) using a detection device, with the self-propelled mobile device determining the position of the detected obstacles (2, 3) on the map of the surroundings (10), classifying the obstacles (2, 3) as surmountable or insurmountable, and passing over an obstacle classified as surmountable (2) and / or going around an obstacle classified as insurmountable (3), characterized in that obstacles near the door are classified as the door threshold and obstacles far from the door are classified as furniture. 2.A method according to claim 1, wherein the classification of obstacles (2, 3) is performed based on existing map data obtained through scanning.

3. A method according to claim 2, wherein the classification is performed by comparing information from scanning with information obtained from obstacle detection (2, 3).

4. A method according to any of the preceding claims, wherein the classification of obstacles (2, 3) is performed before an attempt to cross over the obstacle (2, 3).

5. A method according to any of the preceding claims, wherein the self-propelled mobile apparatus autonomously identifies spaces (1a-1i) as such for classification based on information from its scanning. 6.A method according to one of the preceding claims, wherein obstacles near the door are crossed and obstacles far from the door are bypassed, before making an attempt to cross over obstacles far from the door.

Images